Repository containing code (Transformers_reticulate.R) used in my ESMARConf presentation and some example data that could be used to test out the code
The data folder contains:
- Training and validation data from the SDPRA 2021 Shared Task. This dataset of 35000 article abstracts are labelled by domain (see https://data.mendeley.com/datasets/njb74czv49/1)
- Training and testing data generated from a recent literature search I conducted - texts are labelled as relevant or irrelevant to the hunting or persecution of bat species
UPDATE: Since ESMARConf presentation has been recorded there has been a update to the transformers library which means the trainer module can no longer be used in R. I will be uploading a workaround in due course - watch this space!